By name: clinical-data-management description: | Clinical trial data management, preparation, and analysis methodologies, with SOPs for interim analysis, cohort splitting, and exploratory data analysis. Includes real-world digital health trial examples.
Triggers: "data preparation", "interim analysis", "cohort split", "clinical data management", "trial data analysis", "data source utility", "exploratory data analysis", "digital health trial"
Provides comprehensive SOPs for managing clinical trial data, from source utility planning through interim analysis reporting, with examples from digital health remote monitoring trials (heart failure, cancer care).
Clinical Data Management
Quick Reference
| Phase | SOP/Example | Key Activities | Reference |
|---|---|---|---|
| Interim Analysis | SOP | Protocol, statistical stopping rules, reporting | SOP |
| Cohort Splitting | Data Source Utility Plan | Training/validation split, stratification | SOP |
| Exploratory Analysis | Study Data Explore/Develop | Data profiling, feature exploration, visualization | SOP |
| Trial Design Example | Heart Failure | AIMPower trial - remote monitoring protocol | Example |
| Trial Design Example | Cancer Care | CART trial - remote monitoring in oncology | Example |
When to Use
Use this skill when you are involved in the lifecycle of clinical trial data, specifically:
- Planning Data Strategy: Defining how data sources will be utilized and cohorts defined (Data Source Utility).
- Exploratory Phase: Conducting initial data profiling, quality checks, and hypothesis generation (Study Data Explore/Develop).
- Interim Monitoring: Performing formal interim analyses to assess safety, efficacy, or futility during an ongoing trial.
- Trial Design: Structuring data management plans for digital health interventions (e.g., remote monitoring).
- Predictive Modeling: Preparing clinical datasets for machine learning, including rigorous training/validation splitting.
How to Use
- Identify the Current Phase: Determine if you are in the planning, exploration, or monitoring phase of the clinical trial.
- Select the Appropriate SOP: Use the Quick Reference table to find the Standard Operating Procedure (SOP) that matches your current task.
- Review Real-World Examples: If designing a new protocol, especially for digital health, consult the AIMPower (Heart Failure) and CART (Cancer) examples for structural guidance.
- Apply Statistical Rigor: Use the linked Shared Resources to ensure appropriate statistical tests are selected for your analysis plan.
Data Preparation SOPs
This section details the Standard Operating Procedures for critical data management tasks.
Interim Analysis Reporting
File: references/interim-analysis-report.md
Guidelines for conducting and reporting interim analyses. This SOP covers:
- Establishing the Data Safety Monitoring Board (DSMB) charter.
- Defining statistical stopping rules (e.g., O'Brien-Fleming boundaries).
- Structuring the interim analysis report to maintain blinding where necessary.
Data Source Utility & Cohort Splitting
File: references/data-source-utility-cohort-split.md
Methodology for defining data utility and splitting cohorts for analysis. This SOP covers:
- Data Source Utility Plan (DSUP): Documenting the provenance and intended use of each data source.
- Cohort Splitting: Strategies for creating training, validation, and test sets, ensuring balanced stratification of key clinical variables.
Study Data Exploration & Development
File: references/study-data-explore-develop.md
A framework for Exploratory Data Analysis (EDA) in clinical research. This SOP covers:
- Data profiling and quality assessment (missingness, outliers).
- Univariate and bivariate analysis to understand distributions and relationships.
- Feature engineering and selection for downstream analysis.
Digital Health Trial Examples
Real-world examples of clinical trial protocols focusing on digital health and remote monitoring.
Heart Failure Remote Monitoring (AIMPower)
File: references/trial-protocol-aimpower-heart-failure.md
An example protocol for the AIMPower trial, demonstrating:
- Integration of remote monitoring devices in heart failure management.
- Data collection schedules and patient compliance tracking.
- Endpoints related to readmission reduction and quality of life.
Cancer Care Remote Monitoring (CART)
File: references/trial-protocol-cart-cancer-remote-monitoring.md
An example protocol for the CART trial, illustrating:
- Remote symptom monitoring for cancer patients undergoing treatment.
- Alerting algorithms for clinical decision support.
- Implementation of patient-reported outcomes (PROs) in a digital workflow.
Shared Resources
- Statistical Test Selection Guide: Decision trees and criteria for choosing appropriate statistical tests for clinical trial data analysis. Essential for validating the statistical analysis plan (SAP) associated with these data management activities.
References
- SOPs: Derived from standard clinical data management practices and adapted for digital health contexts.
- Trial Examples: Based on the AIMPower and CART study protocols.